ASR model trained on 6.33 hours of Ikema field data achieves 15% character error rate and reduces transcription time and cognitive load.
Setup In our experiments, we train automatic speech recognition (ASR) models on the newly devel- oped Ikema speech dataset
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Automatic Speech Recognition for Documenting Endangered Languages: Case Study of Ikema Miyakoan
ASR model trained on 6.33 hours of Ikema field data achieves 15% character error rate and reduces transcription time and cognitive load.